With the advent of second generation sequencing (SGS), for the first time there is a truly viable possibility of sequencing a substantial portion of an individual cancer patient's genome within a short time period and at relatively low cost, thus detecting mutations that can have prognostic or predictive value, or can serve as a fingerprint for tumor follow-up in a particular patient. However, there is still a missing link in providing a truly reliable identification of mutation fingerprints in patient tumor samples, as clinically-relevant mutations in tumors with heterogeneity, stromal contamination or in bodily fluids is problematic due to admixture with wild type alleles and can still be missed. And yet, the clinical significance of identifying these low-level mutation fingerprints is major in several situations as frequently these are the mutations that confer resistance, offer prognostic and predictive information and that would be useful for treatment follow-up. Unfortunately the new sequencing technologies 'lose steam' when it comes to detecting low-level mutations, and for SGS currently it's either deep sequencing or high-throughput capability, but not both. Thus integration of SGS with clinical practice cannot be effectively exploited. We developed Co-amplification at Lower Denaturation temperature (COLD-PCR), a new form of PCR that amplifies preferentially the minority alleles from mixtures of wild type and mutation-containing sequences, irrespective of where the mutation lies, providing a strong enrichment of the mutated sequences during PCR. We propose to establish massively-parallel COLD-PCR to enrich mutant sequences prior to their screening via SGA (Illumina), thus enabling 'deep' sequencing while also retaining high-throughput capability. To enable massively-parallel COLD-PCR, a micro-fluidic device that dispenses DNA and PCR reagents within individual nano-droplets (RainDance) prior to PCR will be adapted to perform COLD-PCR in millions of separate nano- reactions simultaneously. The novel combination of technologies will be used to identify mutational fingerprints in tumors from 20 colon cancer patients, including low-level mutations, and then follow these fingerprints in plasma in the course of radio-chemo-therapy, to provide a molecular surrogate to therapy response. The proposed use of Novel Transformative Emerging Technologies is also applicable to other types of cancer and provides a solution bridging the gap in technology and enabling SGS to be applied to clinical oncology practice. Therefore relevance to Public Health is high.